Interference detection and classification in integrated sensing and communication (ISAC) systems is a critical challenge for 6G networks, as it directly impacts system performance and reliability. In this paper, we aim to address this challenge by employing machine learning (ML) techniques, Random Forest, and XGBoost under varying signal-to-noise ratio (SNR) conditions by utilizing radio frequency fingerprints of the interference signal. Our results demonstrate that XGBoost outperforms Random Forest in terms of accuracy, macro-average, and weighted-average metrics as well as other classification metrics. To enhance the interpretability of ML models used, we leverage Explainable AI (XAI) tools, including SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME), to provide insight into the decision-making process of the models that are of paramount importance for AI applications in sixth-generation (6G) of wireless networks. These tools reveal the influence of individual features on the classification of each type of interference, offering a deeper understanding of the underlying patterns. By combining robust ML methods with XAI, this paper not only makes enhancements for interference classification in ISAC systems but also provides actionable insight for designing transceivers for ISAC systems, and tailored solutions for interference cancellation in ISAC systems, paving the way for more reliable and interpretable solutions in future 6G networks.

Interference is a major challenge in integrated sensing and communication (ISAC) systems, and accurate interference classification is critical for ensuring reliable system design. We generate four types of ISAC signals, three of which incorporate self-interference, mutual interference, and clutter interference. We employ three deep learning architectures, including deep neural network (DNN), convolutional neural network (CNN), and long short-term memory (LSTM) models, for interference detection and classification. We compare their performance under different signal-to-noise ratio (SNR) conditions, focusing on both classification accuracy and computational complexity. Simulation results indicate that the CNN outperforms both the DNN and LSTM models in terms of accurate interference classification, albeit at a higher computational complexity. In contrast, the DNN exhibits the lowest complexity but compromises classification accuracy, particularly under low SNR conditions. The LSTM model provides the most balanced performance, effectively balancing classification accuracy and computational complexity.

In this paper, we propose a practical device-free localization and tracking system for mobile objects located in enclosed spaces. The localization is performed by exploiting the RF fingerprints of the object in real time. We investigate three approaches including minimum distance (MD), k-nearest neighbors (KNN), and convolutional neural network (CNN) to obtain centimeter-level localization and tracking. We develop an experimental setup in the laboratory for the proof-of-concept of the proposed methods. This study opens new research directions in the domain of device-free localization and tracking, and offers a scalable and efficient solution for future industrial needs.

In this study, we propose two novel approaches to mitigate cross-link interference (CLI) in a network-controlled repeater (NCR)-assisted integrated sensing and communication (ISAC) network. These approaches include base station (BS) sensing signal transmission (BSST) and an NCR sensing signal transmission (NSST). We implement a power constraint algorithm at the BS to optimize the system performance while ensuring communication reliability and meeting sensing service requirements. Additionally, we investigate the user mobility under the CLI scenarios with different NCR deployments. Specifically, we analyze the NCR connectivity to single and multiple BSs in order to evaluate the received signal reference power and signal-to-interference-plus-noise ratio-based handover (HO) strategies, focusing on radio link failure (RLF), and handover probability. The results show that deploying an NCR-assisted single BS achieves sum rate gains of up to 58.3% and increases in the probability of target detection by 60% compared to networks without assistance. Furthermore, a reduction of RLF by 49.1% is observed in the ISAC networks. These results highlight the effectiveness of NCRs in mitigating CLI and improving the system performance under diverse mobility conditions

In this study, we propose an orthogonal frequency division multiplexing (OFDM) based scheme to achieve interference-free backscatter communications (BC) in a symbiotic radio system. In this scheme, the backscatter device shifts the primary signal, i.e., the OFDM symbols transmitted from a base station, in the frequency domain to transmit its information. Symbiotically, the base station (BS) empties specific subcarriers within the band so that the received signals from the backscatter device and the primary signal are always orthogonal to each other. To address the channel estimation challenge for the signals arriving from the backscatter device, we consider a non-coherent detector for obtaining the information from the backscatter signal at the receiver. We derive the bit-error rate performance of the detector theoretically. Through the comprehensive simulations, we show that the proposed approach achieves a lower bit-error rate up to 10−4at 30 dB with BC by eliminating direct link interference.

In this paper, a novel, yet efficient, multi-user scheduling scheme based on mode selection and power allocation is presented for 5G and beyond networks. The proposed scheme schedules the users in co-existence (CE) and non coexistence (NCE) modes to maximize the connectivity under signal separability and reliability constraints. In addition, two multi-user scheduling mechanisms; check requirements before scheduling (CRBS) and check requirements after scheduling (CRAS), are proposed for practical scenarios to study the impact of quality of service (QoS) on multi-user scheduling in terms of users' reliability requirements. Analytical results show that CRBS outperforms CRAS and orthogonal multiple access (OMA) schemes in terms of connectivity and system throughput with a complexity trade-off.